Population heterogeneity in Plasmodium vivax relapse risk

A key characteristic of Plasmodium vivax parasites is their ability to adopt a latent liver-stage form called hypnozoites, able to cause relapse of infection months or years after a primary infection. Relapses of infection through hypnozoite activation are a major contributor to blood-stage infections in P vivax endemic regions and are thought to be influenced by factors such as febrile infections which may cause temporary changes in hypnozoite activation leading to ‘temporal heterogeneity’ in reactivation risk. In addition, immunity and variation in exposure to infection may be longer-term characteristics of individuals that lead to ‘population heterogeneity’ in hypnozoite activation. We analyze data on risk of P vivax in two previously published data sets from Papua New Guinea and the Thailand-Myanmar border region. Modeling different mechanisms of reactivation risk, we find strong evidence for population heterogeneity, with 30% of patients having almost 70% of all P vivax infections. Model fitting and data analysis indicates that individual variation in relapse risk is a primary source of heterogeneity of P vivax risk of recurrences. Trial Registration: ClinicalTrials.gov NCT01640574, NCT01074905, NCT02143934.


Fig B
Model fit of the constant relapse rate, temporal heterogeneity, the population heterogeneity, and temporal and population heterogeneity model to the first recurrence time in the PNG data. The shaded areas are the 95% confidence regions from the data. For the parameters of the model fit see Tables A, B, C and D in S1 Tables.

Fig C
Fit of models 1 to 3 to the time to first P vivax infection by PCR in the PNG data grouped by village. All villages were fit simultaneously with the same drug washout time distribution, the rate of new infections and relapses were allowed to vary between villages. The lines indicate the model fit and the shaded area the 95% confidence region from the data. For the parameters of the model fit see Tables E, F and G in S1 Tables.

Fig D
Fit of the temporal and population heterogeneity model to the first P vivax infection by PCR in the PNG data grouped by village. All villages were fit simultaneously with the same drug washout time distribution, the rate of new infections and relapses were allowed to vary between villages. The lines indicate the model fit and the shaded area the 95% confidence region from the data. For the parameters of the model fit see Table H in S1 Tables.

Fig E
Association between time to first recurrence and time from first to second recurrence in the Thailand-Myanmar data. Different symbols represent different studies. (B) The symbols represent the different drugs. The Spearman correlation between time to first recurrence and time from first to second recurrence can be found in Table K in S1 Tables.

Fig F
Contribution to recurrences for all patients in the Thailand-Myanmar data. This figure shows which percent of recurrences is caused by which percent of the population. Each dot represents the number of recurrences from 14 to 0 (from left to right), i.e., the first dot represents the percent of the population with at least 14 recurrences (x-axis) and the percent of recurrences caused by the patients with at least 14 recurrences (y-axis). The 20% of the population with the highest number of recurrences cause almost 85% of all recurrences (gray lines). This figure includes all patients, i.e., those who were treated for blood-stage infections only and those who were treated for both blood-and liver-stage infections. For the contribution to relapses for patients who were treated only for blood-stage infections see    The shaded areas are the 95% confidence regions from the data. Abbreviations: AS artesunate, CHQ chloroquine, CHQ/PMQ chloroquine and primaquine, DP/PMQ dihydroartemisinin-piperaquine and primaquine, VHX Vivax History study, BPD best Primaquine Dose study. For the parameters of the model fit see Tables N, O, P and Q in S1 Tables.

Fig K
Model fit of the constant relapse rate, temporal heterogeneity, the population heterogeneity, and the temporal and population heterogeneity model to the first and second recurrence times in the Thailand-Myanmar data simultaneously. The shaded areas are the 95% confidence regions from the data. The comparison of models 3 and 4 with the likelihood-ratio test indicates that model 4 fits the data significantly better than model 3 (p-value < 0.0001). Abbreviations: AS artesunate, CHQ chloroquine, CHQ/PMQ chloroquine and primaquine, DP/PMQ dihydroartemisinin-piperaquine and primaquine, VHX Vivax History study, BPD best Primaquine Dose study. For the parameters of the model fit see Tables R, S, T and U in S1 Tables.  Patients with a follow-up less than one year were excluded. The smooth fit to the data is a smooth spline (fit using the "smooth.spline" function in R with 4 degrees of freedom). The number of recurrences per year for the models were calculated from the simulated data as the average number of recurrences over all individuals with time to first recurrence within the same 10-day time interval.